On Fri, 2 Sep 2016, Pedro Bação wrote:
To replicate Greene's example 9.3 (heteroskedasticity tests), in Gretl I
run the script:
smpl Avgexp>0 --restrict
ols Avgexp const Age Income sq_Income Ownrent
With Gretl 1.9.11, everything goes fine: Gretl manages to drop the various
combinations that are redundant in the auxiliary regression and prints the
When I use Gretl 2016c (Lubuntu 16.04), both versions of the White test
fail with the message "Exact or near collinearity encountered". Am I doing
something wrong in Gretl 2016c, or has a bug creeped into modtest?
Thanks for the report. Indeed, a bug had crept into modtest. This is
now fixed in git, and fixed snapshots will follow shortly.